Particle transport and clustering in stratified turbulent flows

To model the growth of particles by sticking or coagulation, we have developed and tested two conceptually different approaches, which are both implemented in the publicly available Pencil Code. In addition to coagulation, we also include condensation. Both approaches work in the presence of turbulence, which is believed to play a crucial role in cloud cloud droplet growth, for example. In one approach the particles are modelled as a continuous fluid, which is conceptually easier. The other approach uses what is called superparticles or superdroplets, where one superdroplet represents many other ones that would be too expensive to include individually. This approach is now found to be superior over the other one in terms of computational performance. Next, we have considered the clustering of particles by turbulence. It is known that in inhomogeneous turbulence, particles cluster toward the point where the turbulent kinetic energy is a minimum. This process is controlled by two turbulent transport processes; turbophoresis and turbulent diffusion, which together determine the spatial distribution of the particles. If the turbulent diffusivity is assumed to scale with turbulent velocity, as is the case for homogeneous turbulence, the turbophoretic coefficient can be calculated. For the above assumption, it is found that the turbophoretic coefficient has a non-monotonic behavior. This is in contradiction to the classical approximation by Reeks, and it highlights the importance of developing a new description of the turbophoretic effect that is applicable for all particles sizes. We have also studied the mechanism of radiation-induced ignitions that can cause dust explosions. This effect is of great importance for health and safety. In particular, we have studied the effect of particle clustering on the length scale on which radiation causes ignition. We found that the particle clustering results in the formation of small-scale clusters with a high concentration of particles exceeding the mean concentration by 2-3 orders of magnitude. We show that the radiative penetration length strongly increases due to particle clustering. Such strong radiative clearing effect plays a key role for the understanding of the mechanism of multi-point radiation-induced ignitions in dust explosions, and this effect is also of a great importance in the atmospheric and astrophysical turbulence. It has been known for a long time that hydrogen in the gas phase tends to inhibit gasification of char at low and intermediate temperatures. At higher temperatures, however, there are indications that hydrogen may speed up gasification. The mechanisms behind these effects are currently not understood. We have used a newly developed detailed chemical kinetics model for char to study the mechanisms behind the hydrogen inhibition and speed-up of char gasification. For conditions assumed in this work, it is found that for T < 2000 K hydrogen inhibits the heterogeneous reactions, while for T > 2000 K hydrogen in the gas phase speeds up the char conversion. By studying the species reaction rates together with the individual rate of every heterogeneous reaction, the reasons for hydrogen influence on char gasification are explained for a wide range of different temperatures. The effect of turbulence on the heterogeneous (solid?fluid) reactions of solid particles is studied numerically with Direct Numerical Simulations (DNS). We found that, due to the clustering of particles by the isotropic turbulence, the overall reaction rate is entirely controlled by the turbulence for large Damköhler numbers. The particle clustering significantly slows down the reaction rate for increasing Damköhler numbers which reaches an asymptotic limit that can be analytically derived. This implies that the effect of turbulence on heterogeneously reacting particles should be included in models that are used in CFD simulations of e.g. char burnout in combustors or gasifiers. Such a model, based on the chemical and turbulent time scales, is proposed for the heterogeneous reaction rate in the presence of turbulence.

The project aims at resolving issues related to small particles in a turbulent fluid with temperature stratification. This includes, in particular, the effects of turbulent thermal diffusion, small-scale particle tangling clustering and turbulent thermop horesis. These effects are important for a surprisingly diverse set of applications such as formation of clouds, agglomeration of water droplets leading to rain, fouling on super heaters in industrial boilers, and the formation of planets from protoplane tary clouds. The findings of this project will therefore be significant for the understanding of planet formation, the design of power plants utilizing difficult fuels such as biomass and municipal solid waste, and cloud and rain formation. We use the P encil Code both in the fully compressible and the newly developed anelastic configurations, allowing us to access the low Mach number regime.